The objective of this study was to examine the efficacy and tolerability of intermittent dosing of venlafaxine for the treatment of premenstrual dysphoric disorder. One hundred and twenty-four women aged 18 to 45 years, with regular menstrual cycles and who reported significant premenstrual symptoms, were assessed prospectively to confirm their diagnosis of premenstrual dysphoric disorder. Twenty subjects with confirmed premenstrual dysphoric disorder entered a single-blind, placebo phase (1 cycle). Placebo nonresponders (n = 12) received 2 cycles of intermittent (premenstrual) treatment with venlafaxine (75 to 112.5 mg/d). Subjects initiated treatment 14 days before the anticipated onset of menses and discontinued it on the second day of bleeding. Doses could be adjusted after cycle 1 based on subjects' response and tolerability. Response to treatment was assessed based on changes in the Daily Rating Severity of Problems and Premenstrual Tension Syndrome Questionnaire scores from baseline (before the placebo cycle), as well as Clinical Global Impression-Severity scores. Discontinuation symptoms were assessed between treatment cycles, using the Discontinuation-Emergent Signs and Symptoms questionnaire. Eleven subjects concluded 2 cycles of intermittent dosing with venlafaxine. Nine subjects (81.8%) showed satisfactory response based on Clinical Global Impression of < or = 2. Changes in Daily Rating Severity of Problems scores and subscores (depression, physical symptoms, and anger) and in Premenstrual Tension Syndrome Questionnaire scores were significant (P < 0.05 for all comparisons, Wilcoxon tests). Intermittent treatment was well tolerated. This preliminary report suggests that premenstrual use of venlafaxine is an efficacious and well-tolerated treatment for premenstrual dysphoric disorder.